During the last 40 years, debate on the origins of language has featured two theoretically dichotomous positions, one that language is largely innately specified, and another, that language is explicitly learned. Recent advances in our understanding of language development take us beyond assertions that language is either instinctual or learned, and teach us something about the actual steps involved in infants' acquisition of a particular language. The data show that biological predispositions and the ability to acquire information through exposure to language are inextricably intertwined. More important than the "interactionist" view this reflects are the details of the findings themselves, revealing how infants move from one level to another, acquiring novel information from exposure to language, using strategies we had not predicted. Data from this laboratory show that in the earliest periods of development, infant perception is structured in a way that greatly assists speech perception, demonstrating a rich initial structure that supports language. Moreover, our cross-language studies show that once experience occurs, infants display an extraordinary ability to acquire the unique properties of a specific language. We show that exposure to a particular language alters infant perception early and in an interesting way. Perception is dynamically restructured as infants "map" the phonetic parameters of their native language. These new data prompted Kuhl to articulate a new theory called the Native Language Magnet theory. It has begun to have an impact within speech and language, as well as in broader areas, including child development, neuroscience, neurobiology, and computational modeling. The theory acts as the background framework for the studies in this proposal. Four converging lines of research are proposed to test the theory and further advance our knowledge of infant speech development: (a) developmental change in infant speech perception, examining the transition in speech perception that occurs between 6 and 12 months; (b) brain correlates (MMN) of speech perception in 6-12 month old infants, (c) analysis of language input to infants, examining the phonetic content of infant-directed speech in detail, and (d) plasticity for phonetic information in infancy, studies that examine the impact of exposure to a foreign-language to assess whether, when, and how infants' sensitivity to new phonetic information changes with age and linguistic experience. The four converging lines of research should produce data that not only address specific theories of speech and language development, but more general theories of the interface between biology and culture.